The argument “the water use for beef is exorbitantly high” (short version: “a gazillion litres of water go into a single burger”) is fairly common - but does it actually hold up?

A rant - on a whim, and not finished.

Regular readers here will have noticed that I’m a vehement advocate of pasture farming. In that case the average rainfall onto a pasture would also be counted towards the “gazillion litres of water"(1). Yet it would have rained on that pasture with or without cattle - only, without cattle the pasture would have markedly lower biodiversity or ecological value. So it’s not very honest to count that water, especially since a cow only uses it temporarily before excreting it again on the very same spot.

Grass has to be grazed off now and then to form its roots properly and to renew itself. Various insects need the dung that cattle (and other ruminants) leave behind to lay their eggs, and as food for their offspring.

The numbers that count

But what trumps all of these arguments is an overview by the German Federal Environment Agency on “water resources and their use” (I’m looking at the 2022 version as I write this, and yes, it’s in german language, but you can understand the graph contained within it):

Pie-chart that shows the numbers discussed in the article
The graph from the discussed publication ('Landwirtschaft' means 'agriculture', 'Öffentliche Wasserversorgung' means 'public water supply')

Listed there: the total water withdrawal for the year, made up of surface water, groundwater etc. These figures contain no rainwater. Agriculture (and that’s all of it, plant and animal alike) “consumes"(2) a mere 2.5% of the water withdrawn, while the public water supply - which includes household use - accounts for 29.8% of the water withdrawn.

In volume terms (once you net out exports and imports) Germany is more than self-sufficient (3) in beef(4) - which means the water use for cattle farming is a fraction of agricultural use, and therefore less than 2.5% of the total.

Rain doesn’t count as a water withdrawal, but since the rain would fall anyway - and we don’t tally it into the water use for street cleaning either - I regard the “virtual water” in the burger, those very gazillion litres, as a completely inadmissible basis for comparison.

If agriculture accounts for only 2.5% of the water withdrawn, and animal agriculture only a fraction of that, and bringing up ruminants only a fraction of that, then we have distinctly better places to cut water use than the production of our food - and, on top of that, of the most nutrient-dense foods known.

Bottom line

Twelve times as much water goes into the public water supply as into agriculture. Anyone worried about the water in their burger should first rip out their own bathtub, shower in the rain and worry less about the non-productive grassland around their house: the lawn sprinkler gets questioned far less often.

In other words: the “gazillion litres” yardstick isn’t watertight: it is Good that so much water goes into a burger - anything else would mean drought and desertification.

More on pasture farming, especially on desertification

Cattle herd on pasture, in front of trees. One white animal among an otherwise black herd
Pasture-Raised Ruminants

Cattle get a lot of blame - how the farming method can turn that picture around

A cow on grass. In the background, more cattle lie partially hidden in tall grass.
Allan Savory: What Holds Up, What Doesn't

In the first grazing article I flagged some dissent about Allan Savory. Turns out it’s not one but two objections.

  1. But even in the case of industrially raised livestock I can’t identify any other source than rain - so the argument holds there as well
  2. I write “consumes” even though the water is of course still available afterwards. A generous choice of words, but ultimately it applies to almost any water that isn’t subjected to electrolysis.
  3. Federal Office for Agriculture and Food (Bundesanstalt für Landwirtschaft und Ernährung): Meat supply balance 2024 (german language)
  4. The self-sufficiency rate is a pure quantity balance of the end product (beef produced vs. consumed) and says nothing about where the feed came from. But that caveat shrinks on closer inspection: around 90% of the feed used in Germany is domestic; the import gap is almost entirely protein feed (soy, ~30% self-sufficiency), and that soy goes mostly to pigs and poultry. Ruminants cover their needs overwhelmingly from domestic grassland and - for protein, increasingly - from domestic rapeseed meal, a by-product of rapeseed-oil production (also check the mention of press-cake in Animal Suffering). Pasture cattle need no imported feed anyway. But even the imported soy grows mostly in the rain. Source: German Farmers’ Association (german language)